History, computing and datafication: an old challenge

The rise of computing, networks and data since 1945.

• 1959 (dealing with too much data): François Furet and Adeline Daumard. ‘Méthodes de l’Histoire sociale: les Archives notariales et la Mécanographie’. Annales ESC, 14(4).

• 1961 (cross-referencing two datasets): Paul Garelli and Jean-Claude Gardin. ‘Étude Par Ordinateurs Des Établissements Assyriens En Cappadoce’. Annales ESC, 16(5).

Datafication?

• More than digitization
• The transformation of digitized (and born digital) material into data that can be computed and hence analysed
• The ideology that those data can represent better the society (or the past) than analog (traditional) interpretations

Mayer-Schönberger, Viktor, and Kenneth Cukier. Big Data: A Revolution That Will Transform How We Live, Work, and Think. Houghton Mifflin Harcourt, 2013.

At the root of datafication: (personal) computer and networks

• 1980s: rise of personal computing
• 1990s: internet / web
• 2000s: social media / platformization of the web (i.e. big data)
• 2010s: AI (Machine/deep learning)

History and cognitive delegation

• Some historians seized those evolutions from the very beginning
• Emergence of Digital Humanities and Digital History since the mid-2000s
• Many projects that uses AI

Ex:

• Trésor des Chartes
• Transkribus

The difficult question of the historian’s computing skills (French case)

• End of the 1970s: French Medievists rise the question of teaching computing in History
• 1986: Jean-Philippe Genet: Genet, Jean-Philippe. ‘Histoire, Informatique, Mesure’. Histoire & Mesure, 1(1), 1986.
• 1993: Genet, Jean-Philippe. ‘La Formation Informatique Des Historiens En France: Une Urgence’. Mémoire Vive, no. 9.
• 2011: Ruiz, Émilien, and Franziska Heimburger. ‘Faire de l’histoire à l’ère Numérique : Retours d’expériences’. Revue d’histoire Moderne et Contemporaine, n° 58-4bis.
• 2019: dhiparis. ‘Enseigner le numérique aux historien·ne·s – perspectives internationales #dhiha8’. Digital Humanities à l’Institut historique allemand.

for non-French examples:

• Salmi, Hannu. What Is Digital History?. Polity, 2020.

The datafication of History

Mass digitization of primary (and secondary) sources + cheap computing + network

• Transition from primary sources to dataset
• The emergence of the big data of the past

Platformization of the web

• The emergence of new primary sources (ex: social media)

New challenges: Preservation

• Rosenzweig, Roy. ‘Scarcity or Abundance? Preserving the Past in a Digital Era’. The American Historical Review, 108(3), June 2003.)

New challenges: the Democratization of history?

• Democratization through accessibility
• No democratization of methodology

Revisiting old primary sources through digitization

The case of digitized Newspapers

• Newspapers as the most digitized kind of primary sources
• Renewal of media history
• Numerous projects based on digitised newspapers (= form of datafication)
  • Oceanic Exchanges, Numapress, Impresso…

The Impresso Project

Swiss project (EPFL, Université de Lausanne, University of Zürich) with Luxemburgish partnership (C²DH)

Bunout, Estelle, et al., eds. Digitised Newspapers – A New Eldorado for Historians? Tools, Methodology, Epistemology, and the Changing Practices of Writing History in the Context of Historical Newspapers Mass Digitization. De Gruyter Oldenbourg, 2022.

Taking into account the datafication phenomenon

Big data of the Past and the flaws of Big data

Big Data (of the past) is a social construct as all data.

Boyd, Danah, and Kate Crawford. ‘CRITICAL QUESTIONS FOR BIG DATA: Provocations for a Cultural, Technological, and Scholarly Phenomenon’. Information, Communication & Society, 15(5), 2012.

Big data of the past and the lamp post effect

- Claire Lemercier and Claire Zalc, Quantitative Methods in the Humanities. An Introduction, London: University of Virginia Press, 2019.

- Milligan, Ian. ‘Illusionary Order: Online Databases, Optical Character Recognition, and Canadian History, 1997-2010’. Canadian Historical Review, 94(4), 2013.

- Putnam, Lara. ‘The Transnational and the Text-Searchable: Digitized Sources and the Shadows They Cast’. The American Historical Review, 121(2), 2016.

Big data of the past and AI

• AI today is mostly linked to machine learning / deep learning
• learning => training
• training => big data
• if training datasets are biased, so is any AI-based system

Exploring new sources: the example of social media

• New Sources for future historians
• Today’s sources for historians (sociologists, linguists…) interested in the link between history and collective (cultural) memory

#ww1: echoes of the Centenary of the Great War on Twitter

The question of distant reading.

Fragility

Well. Elon Musk.

The diversity of the echoes of the past

Most citizens are confronted to the (historical) past in a non-academic / secondary education context.

• The most striking example is video games

The non-existence of a digital native generation

• There’s no digital native generation
• Maybe Facebook (or Snapchat, or Instagram, or Whatsapp, etc – depending on places and time) native generation

We should not assume that our students are better fitted to the datafied world we are living in.

basics are still valid…

• The importance of primary sources
• The importance of methodology
• The importance of reading

…but to be updated

• From close reading to distant reading to scalable reading
• Critical assesment of primary sources

Haber, Peter. ‘Digital Past: Geschichtswissenschaft im digitalen Zeitalter’. Digital Past, Oldenbourg Wissenschaftsverlag, 2011.

Clavert, Frédéric, and Andreas Fickers. ‘On Pyramids, Prisms, and Scalable Reading’. Journal of Digital History, no. jdh001, 2021.

Thank you!

Prez available online: https://inactinique.github.io/